New Evidence on the Puzzles. Results from Agnostic Identification on Monetary Policy and Exchange Rates.



4.2 Foreign Monetary Policy Shocks

For a foreign monetary policy shock, figure 14 contains the key results, while
figures 15 to 20 contain further details.

The first line of figure 14 shows the impulse response of the real exchange
rate and should be compared to the theory picture 3. The second line shows
the posterior probability for that impulse response function to be positive in
any given month. The results for nominal and real exchange rates are very
similar throughout: we thus only show the latter.

The results show that, again, there is a gradual response only, i.e. a
delayed overshooting, and that there is considerable probability for the ex-
change rate to appreciate rather than depreciate on impact. The results here
are rather robust across countries. This is in contrast to Grilli and Roubini
(1995, 1996) who find the exchange rate puzzle for Germany but not for the
UK.

Figure 15 compares the response of the real exchange rates for the sign
restriction approach in the GR as well as the BIG specification to the results
from the recursive Grilli-Roubini identification. The Grilli-Roubini identifi-
cation leads to very different results, depending on the country pair, whereas
the results for the sign restriction approach look considerably more alike.

Figures 16 and 17 provide further comparisons between GR and the origi-
nal Grilli-Roubini identification on the distribution of the sign of the response
for the real exchange rate as well as the price level. For our specification and
on impact, the posterior probability for a positive exchange rate response is
somewhere between 30% and 60%. This probability then rises and plateaus
somewhere near 70% to 90% within three years. There is more variation
in the results regarding the sign distribution for the original Grilli-Roubini
identification. There also is a considerably large price puzzle. We believe
that this shows the advantage of using the sign restriction methodology as
opposed to their original identification strategy, when applying it to this

20



More intriguing information

1. Modelling the health related benefits of environmental policies - a CGE analysis for the eu countries with gem-e3
2. On the Existence of the Moments of the Asymptotic Trace Statistic
3. The name is absent
4. Partner Selection Criteria in Strategic Alliances When to Ally with Weak Partners
5. The name is absent
6. TOWARDS THE ZERO ACCIDENT GOAL: ASSISTING THE FIRST OFFICER MONITOR AND CHALLENGE CAPTAIN ERRORS
7. Confusion and Reinforcement Learning in Experimental Public Goods Games
8. Are Japanese bureaucrats politically stronger than farmers?: The political economy of Japan's rice set-aside program
9. TOWARD CULTURAL ONCOLOGY: THE EVOLUTIONARY INFORMATION DYNAMICS OF CANCER
10. The name is absent